geoskill-post-fire-recovery

火后植被恢复监测 | Post-Fire Recovery

Monitors post-fire vegetation dynamics in two steps: first, dNBR is used to determine burn severity; second, the post-fire multi-temporal NDVI time series is used to track the recovery trajectory. It suits burned-area loss assessment, vegetation recovery monitoring and post-disaster ecological restoration planning.

Core algorithm:

• Burn severity (dNBR): the normalized burn ratio NBR=(NIR−SWIR)/(NIR+SWIR) is differenced as dNBR=NBR_pre−NBR_post. Following the key thresholds of USGS / Key et al. 2006, five severity classes are defined: unburned (<0.10) / low (0.10–0.27) / moderate_low (0.27–0.44) / moderate_high (0.44–0.66) / high (≥0.66).
• Recovery slope: per-pixel linear slope of the post-fire NDVI series; positive values indicate ongoing recovery.
• Recovery year: the epoch index at which NDVI first returns to the pre-fire baseline × target (default 0.95); pixels not recovered within the observation period are recorded as −1.
• Recovery trajectory: spatially averaged NDVI per epoch (optionally restricted to the burned area).

The --synthetic mode generates physically consistent scenes with varying burn severity and recovery rates (offline).

Dependencies / 依赖

pip install numpy rasterio scipy

Usage / 使用方法

Basic usage (synthetic data, offline)

python geoskill-post-fire-recovery.py --bbox 118.0 34.0 119.0 35.0 --synthetic --n-dates 6 --output-dir ./output

Example 1: adjust the recovery target ratio

python geoskill-post-fire-recovery.py \
    --bbox 118.0 34.0 119.0 35.0 \
    --synthetic --n-dates 6 --recovery-target 0.90 \
    --output-dir ./target90

Example 2: real multi-band imagery

python geoskill-post-fire-recovery.py \
    --input fire_scene.tif \
    --n-dates 6 \
    --output-dir ./real

Input band order: nir_pre / swir_pre / nir_post / swir_post / ndvi_prefire, followed by n-dates bands of post-fire NDVI per epoch.

Output / 输出

Data Source / 数据源 / Source

• Synthetic mode: generated locally, no external data sources
• Real mode: user-provided pre-fire/post-fire multi-band GeoTIFF (e.g., Landsat / Sentinel-2)

Privacy / 隐私声明 / Privacy

• Fully offline by default; no network requests are made
• All computation is performed locally; user data is never uploaded

License / License

MIT

---

---

name: geoskill-post-fire-recovery description: '用 dNBR（NIR/SWIR 差分归一化烧伤比）判定五级烧伤严重度，结合火后多期 NDVI 恢复曲线估算恢复轨迹、恢复斜率与恢复年限，输出严重度 GeoTIFF、恢复轨迹 JSON 与恢复年限栅格。Post-fire recovery from dNBR severity and NDVI time series.'

火后植被恢复监测 | Post-Fire Recovery

分两步监测火灾后的植被动态：先用 dNBR 判定烧伤严重度，再用火后多期 NDVI 时间序列追踪恢复轨迹。适用于过火区损失评估、植被恢复监测与灾后生态恢复规划。

核心算法：

• 烧伤严重度（dNBR）：归一化烧伤比 NBR=(NIR−SWIR)/(NIR+SWIR)， 差分 dNBR=NBR_pre−NBR_post。按 USGS / Key et al. 2006 关键阈值分为五级： unburned (<0.10) / low (0.10–0.27) / moderate_low (0.27–0.44) / moderate_high (0.44–0.66) / high (≥0.66)。
• 恢复斜率：火后 NDVI 序列的每像元线性斜率，正值表示恢复中。
• 恢复年限：NDVI 首次回到火前基线 × target（默认 0.95）的期号；观测期内 未恢复记 −1。
• 恢复轨迹：逐期空间平均 NDVI（可限定在过火区内）。

支持 --synthetic 模式生成含不同烧伤严重度与恢复速率的物理一致场景（离线）。

依赖

pip install numpy rasterio scipy

使用方法

基本用法（合成数据，离线）

python geoskill-post-fire-recovery.py --bbox 118.0 34.0 119.0 35.0 --synthetic --n-dates 6 --output-dir ./output

示例 1：调整恢复目标比例

python geoskill-post-fire-recovery.py \
    --bbox 118.0 34.0 119.0 35.0 \
    --synthetic --n-dates 6 --recovery-target 0.90 \
    --output-dir ./target90

示例 2：真实多波段影像

python geoskill-post-fire-recovery.py \
    --input fire_scene.tif \
    --n-dates 6 \
    --output-dir ./real

输入波段顺序：nir_pre / swir_pre / nir_post / swir_post / ndvi_prefire，其后 n-dates 个波段为火后各期 NDVI。

输出

数据源 / Source

• 合成模式：本地生成，无外部数据源
• 真实模式：用户提供火前/火后多波段 GeoTIFF（如 Landsat / Sentinel-2）

隐私声明 / Privacy

• 默认完全离线运行，不发起任何网络请求
• 所有计算在本地完成，不上传用户数据

License

MIT